An AI device currently awaiting approval by the FDA may help identify newborns at risk for aggressive posterior retinopathy of prematurity (AP-ROP), a condition that is difficult to diagnose and can lead to vision loss if left untreated.

In a recent study, 947 newborns in nine neonatal care centers were followed over time, and fundus images from a total of 5945 eye examinations were analyzed both by a deep learning system and a team of expert fundus image graders.

“Artificial intelligence has the potential to help us recognize babies with AP-ROP earlier. But it also provides the foundation for quantitative metrics to help us better understand AP-ROP pathophysiology, which is key for improving how we manage it,” said the study’s lead investigator, J. Peter Campbell, M.D., M.P.H., Casey Eye Institute, Oregon Health and Science University in Portland.

Babies born prematurely are at risk for retinopathy, meaning they have fragile vessels in their eyes that can leak blood and grow abnormally. If left untreated, vessel growth can get worse and cause scarring, leading to detachment of the retina and vision loss. The incidence of ROP each year in the US is about 0.17 percent, and most cases are mild and resolve without treatment.

Upon birth, the eyes of premature babies are screened and closely watched for signs of retinopathy. ROP-related changes occur along a spectrum of severity, however, and AP-ROP can elude diagnosis because its symptoms can be more subtle than those of typical ROP. While AP-ROP was recognized as a diagnostic entity in 2005, there is still significant variation among clinicians about whether eyes show signs of AP-ROP.

“Even the most highly experienced evaluators have been known to disagree about whether fundus images indicate AP-ROP,” Dr. Campbell noted, and previous studies have shown that deep learning can outperform clinicians in detecting subtle patterns in fundus images and classifying ROP.

“It’s important to acknowledge that there is currently no gold standard for diagnosing AP-ROP,” said Grace L. Shen, Ph.D., who manages the retinal diseases program for the Division of Extramural Science Programs at the National Eye Institute. “But having objective, AI-based metrics for detecting AP-ROP is a step in the right direction for this highly vulnerable population of infants.”

The deep learning system used in the study, called the i-ROP DL system, was recently granted breakthrough status by the FDA, which will help accelerate the development and potential approval of the device.